My AI Agent Swarm Works in 3D (Insane to Watch)
Read the Portuguese translation →
Author: Leon van Zyl · Watch on YouTube
Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.
Original language: English
Original video: My AI Agent Swarm Works in 3D (Insane to Watch) · Channel: Leon van Zyl · 00:30 · Watch on YouTube
Topics: AI agents, Task and workflow automation, Software engineering with AI, Open source and software ecosystems
Summary
Section titled “Summary”Leon van Zyl presents a 3D visualization of his coding-agent swarm working on a real project. Each agent runs a coding harness, such as Claude Code or Codex, and the monitors reveal which tools the agents are using. The swarm is connected to a Kanban board and GitHub issues, with links to pull requests. Agents continue working in the background until changes are pulled, tested, and verified.
Key points
Section titled “Key points”- The coding-agent swarm is represented in 3D space.
- The agents are working on an actual project.
- Each agent runs a coding harness, such as Claude Code or Codex.
- The monitors can show which harness an agent is running, including Claude Opus via Claude Code.
- Some agents use other coding harnesses.
- The setup includes a Kanban board.
- The agents are connected to actual GitHub issues.
- Clicking an agent can open its pull request on GitHub.
- Agents work in the background until everything is pulled, tested, and verified.
Chapters
Section titled “Chapters”| Start | Chapter | Summary |
|---|---|---|
| 00:00 | A coding-agent swarm in 3D | Leon introduces his 3D coding-agent swarm and shows that its agents run coding harnesses such as Claude Code and Codex. |
| 00:30 | Kanban and GitHub integration | He explains that the agents connect to Kanban and GitHub issues, link to pull requests, and keep working until changes are pulled, tested, and verified. |
Quotes
Section titled “Quotes”No one said an agent harness should be boring. — 00:00
This is my coding agent swarm represented in 3D space. — 00:00
When I click on any of these, we can actually jump to the actual pull request sitting on GitHub. — 00:30
These agents will keep working away in the background until everything was pulled, tested, and verified. — 00:30
Concepts and entities
Section titled “Concepts and entities”coding agent swarm, Claude Code, Codex, Claude Opus, GitHub, Kanban board
Full transcript (original language)
Section titled “Full transcript (original language)”00:00 No one said an agent harness should be boring. This is my coding agent swarm represented in 3D space. And these guys are working on an actual project. Each of these agents are running a coding harness. So, this could be Claude code, it could be Codex. If I run to any of these monitors, I can actually see like this guy is actually running Claude Opus. So, this is using Claude code. And some of them are using other coding harnesses as well. And this also uses the Kanban board. And these are hooked
00:30 up to actual GitHub issues. When I click on any of these, we can actually jump to the actual pull request sitting on GitHub. And these agents will keep working away in the background until everything was pulled, tested, and verified.